{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6NSCPLZ62MU5B466CVSKAU3DON","short_pith_number":"pith:6NSCPLZ6","schema_version":"1.0","canonical_sha256":"f36427af3ed329d0f3de1564a053637372d2b9f9dcd944ead543645955787db4","source":{"kind":"arxiv","id":"2410.10181","version":2},"attestation_state":"computed","paper":{"title":"Scalable Multi-Domain Adaptation of Language Models using Modular Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Arun Kandoor, Chih-Kuan Yeh, James Laudon, Peter Schafhalter, Shun Liao, Yanqi Zhou","submitted_at":"2024-10-14T06:02:56Z","abstract_excerpt":"Domain-specific adaptation is critical to maximizing the performance of pre-trained language models (PLMs) on one or multiple targeted tasks, especially under resource-constrained use cases, such as edge devices. However, existing methods often struggle to balance domain-specific performance, retention of general knowledge, and efficiency for training and inference. To address these challenges, we propose Modular Domain Experts (MoDE). MoDE is a mixture-of-experts architecture that augments a general PLMs with modular, domain-specialized experts. These experts are trained independently and com"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.10181","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-14T06:02:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"771b5b4457f5edc95a508f3b65304197b3270264bb362d1a8800a4d8a0dc8346","abstract_canon_sha256":"a7029efe750eeb876638d5a268d7777349815d96a60388864da78921e6064621"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:25:02.203660Z","signature_b64":"ZG3Lk+dOEGgN/l1gkHkuzkwVh+aI3CILw6rcUQ72Kg7bxrWr9UKDNc3iDhD94Qknzs8vcERXv9Zgd4SCrPkbDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f36427af3ed329d0f3de1564a053637372d2b9f9dcd944ead543645955787db4","last_reissued_at":"2026-07-05T09:25:02.203220Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:25:02.203220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable Multi-Domain Adaptation of Language Models using Modular Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Arun Kandoor, Chih-Kuan Yeh, James Laudon, Peter Schafhalter, Shun Liao, Yanqi Zhou","submitted_at":"2024-10-14T06:02:56Z","abstract_excerpt":"Domain-specific adaptation is critical to maximizing the performance of pre-trained language models (PLMs) on one or multiple targeted tasks, especially under resource-constrained use cases, such as edge devices. However, existing methods often struggle to balance domain-specific performance, retention of general knowledge, and efficiency for training and inference. To address these challenges, we propose Modular Domain Experts (MoDE). MoDE is a mixture-of-experts architecture that augments a general PLMs with modular, domain-specialized experts. These experts are trained independently and com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.10181","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2410.10181/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.10181","created_at":"2026-07-05T09:25:02.203285+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.10181v2","created_at":"2026-07-05T09:25:02.203285+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.10181","created_at":"2026-07-05T09:25:02.203285+00:00"},{"alias_kind":"pith_short_12","alias_value":"6NSCPLZ62MU5","created_at":"2026-07-05T09:25:02.203285+00:00"},{"alias_kind":"pith_short_16","alias_value":"6NSCPLZ62MU5B466","created_at":"2026-07-05T09:25:02.203285+00:00"},{"alias_kind":"pith_short_8","alias_value":"6NSCPLZ6","created_at":"2026-07-05T09:25:02.203285+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.17359","citing_title":"Learning Transferable Topology Priors for Multi-Agent LLM Collaboration Across Domains","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON","json":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON.json","graph_json":"https://pith.science/api/pith-number/6NSCPLZ62MU5B466CVSKAU3DON/graph.json","events_json":"https://pith.science/api/pith-number/6NSCPLZ62MU5B466CVSKAU3DON/events.json","paper":"https://pith.science/paper/6NSCPLZ6"},"agent_actions":{"view_html":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON","download_json":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON.json","view_paper":"https://pith.science/paper/6NSCPLZ6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.10181&json=true","fetch_graph":"https://pith.science/api/pith-number/6NSCPLZ62MU5B466CVSKAU3DON/graph.json","fetch_events":"https://pith.science/api/pith-number/6NSCPLZ62MU5B466CVSKAU3DON/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON/action/storage_attestation","attest_author":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON/action/author_attestation","sign_citation":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON/action/citation_signature","submit_replication":"https://pith.science/pith/6NSCPLZ62MU5B466CVSKAU3DON/action/replication_record"}},"created_at":"2026-07-05T09:25:02.203285+00:00","updated_at":"2026-07-05T09:25:02.203285+00:00"}